NSCLCMA
NSCLCMA
NSCLCMA
take account of the expanded evidence base and provide the most queries were resolved and the final database entries verified by the responsible
up-to-date and reliable assessment of the role of chemotherapy in trial investigator or statistician.
NSCLC, the NSCLC Collaborative Group has carried out an updated Definition of Outcomes
IPD meta-analysis that examines the role of chemotherapy in seven Overall survival was defined as the time from randomization until death
treatment comparisons.7 In the supportive care setting reported here, by any cause. Patients still alive were censored at the date of last follow-up or
we assessed the role of newer chemotherapy agents and assessed more date last known to be alive.
reliably the effect of chemotherapy in different subgroups of patients. Analysis
Analyses of outcomes, trial groups, and patient groups were (unless
otherwise stated) prespecified in the protocol and carried out on an intention
METHODS to-treat basis; that is, patients were analyzed according to their allocated treat-
ment, irrespective of whether they received that treatment. Analyses of all end
The aim was to assess the effect of supportive care and chemotherapy versus points were stratified by trial, and the log-rank expected number of deaths and
supportive care alone in advanced NSCLC. The meta-analysis followed a variance was used to calculate individual trial hazard ratios (HRs) and overall
detailed and prespecified protocol which set out the objectives, inclusion pooled HRs based on the fixed effect model.9 Thus, the times to death for
criteria for trials, data to be collected and analyses to be carried out. A copy of individual patients were used within trials to calculate the HR, representing the
the protocol is available on request (from S.B.). overall risk of an event for those patients allocated to supportive care and
chemotherapy compared with those allocated to supportive care alone. Results
Inclusion Criteria were also combined using the random effects model to assess the robustness of
To be included in the meta-analysis, trials had to be properly random- the results to the choice of meta-analysis model.
ized, have commenced accrual on or after January 1, 1965, and have completed To examine the potential impact of the treatments used, we predefined
accrual. Trials should have included patients with NSCLC who had received analyses that grouped trials by the type of the chemotherapy regimen used. For
either chemotherapy and supportive care or supportive care alone, that were these analyses, a pooled HR was calculated for each group of trials and for all
unsuitable for surgery or radical radiation therapy. Supportive care was de- trials together. As we specified that we would group those trials using
fined in the individual trials and may include palliative radiotherapy, antibiot- platinum-based chemotherapy with a vinca-alkaloid or etoposide separately
ics, corticosteroids, analgesics, antiemetics, transfusions, and psychosocial from those trials that did not use a vinca-alkaloid or etoposide, the Big Lung
support. Patients should have not received any previous chemotherapy or had Trial (BLT) trial10 was divided into two trials. BLT1 combined cisplatin with a
any prior malignancy. vinca-alkaloid and BLT2 did not use a vinca-alkaloid.
Identification of Trials The relative effects of chemotherapy in different subgroups of patients
To limit publication bias, we included all randomized trials, whether were investigated using similar stratified analyses. Analyses were performed for
published or unpublished. We carried out bibliographic searches of Medline each prespecified subgroup, for example, comparing the effect of treatment
and CancerLit using the Cochrane Collaboration optimal search strategy for and control for males and for females within each individual trial. These results
identifying randomized controlled trials.8 These were supplemented by were then combined to give overall HRs for males and for females. 2 tests for
searching the Cochrane Central Register of Controlled Trials, the abstracts interaction or trend were used to investigate whether there were any substan-
from relevant conferences, the reference lists of identified trials, the bibliogra- tial differences in the effect of chemotherapy between groups of trials or
phies of relevant books, and review articles. The National Cancer Institute subgroups of patients.
Physicians Data Query clinical protocols, United Kingdom Coordinating Results are also presented as absolute differences at 1 year, calculated
Committee for Cancer Research trials register, and the Current Controlled using the overall HRs and the control arm event rate.11 CIs for absolute
Trials metaRegister of trials were also searched to identify unpublished and differences were calculated from the baseline event rate and the HR at the
ongoing trials. All trialists who took part in the meta-analysis were asked to 95% CI boundary values. 2 heterogeneity tests and the I2 statistic for
help to identify additional trials. Initially searches were completed for the inconsistency12 were used to assess statistical heterogeneity across trials. Sur-
period up to and including 2003. These were revised regularly to identify vival curves are presented as simple (nonstratified) Kaplan-Meier curves.13 All
further trials published by our final analyses in September 2007. Where there P values quoted are two sided.
was uncertainty about the eligibility of a trial or particular treatment arms Median follow-up was calculated by the reverse Kaplan-Meier method,
within a trial, this was discussed and resolved by consensus within the Project based on surviving patients and using censoring as the event.
Secretariat and International Advisory Group. We did not search for trials that
used long-term alkylating agents. These were included in the 1995 analyses,
but due to their antiquity, were not included in this update. RESULTS
Data Collection
For trials already included in the 1995 analyses, updated follow-up was A total of 19 potentially eligible trials that had used supportive care and
sought. Most of the trials previously provided mature data and we did not chemotherapy versus supportive care alone were identified. Data for
anticipate much additional information. However, some additional data were two trials14,15 were no longer available and for one trial,16 adequate
received and included in the new analyses. contact with the investigators could not be established.
For new trials, survival and baseline characteristics were sought for all Therefore, 16 trials that randomly assigned 2,714 patients were
patients randomly assigned into each trial. This included date of random- included (Table 1).10,17-29 These represent 84% of patients from all
ization, survival status, and date of last follow-up or death as well as
known randomized trials that compared supportive care and chemo-
information on date of birth, sex, performance status, tumor stage (TNM),
and histological type. therapy with supportive care alone and 65% more data than that
available in 1995. The 16 trials accrued between 32 to 447 patients.
Data Checking Characteristics of these trials are summarized in Table 1. Platinum-
A number of standard checks were applied to all new trials, including based chemotherapy was used in 12 trials (cisplatin in 11 and
checks for missing values and data validity and consistency across variables. To
assess the randomization integrity, we looked for unusual patterns in the
carboplatin in one25) and nonplatinum single agents (etoposide,
sequencing of allocation or imbalances in baseline characteristics between vinorelbine, gemcitabine, and paclitaxel) were used in four trials.
treatment arms. Follow-up of surviving patients was also assessed to ensure Patients’ characteristics for the 2,714 patients across all trials are
that it was balanced by treatment arm and as up-to-date as possible. Any presented in Table 2. Data for age and sex were provided for all trials.
Using platinum ⫹
vinca-alkaloid/
etoposide
RLW 835117 1982-1986 167 Cisplatin 120 P/T 0 0 100 0 73 27 37 37 26 88 12
Vindesine 3
NCIC CTG BR518 1983-1986 150 Cisplatin 120 P/T 0 0 13 85 60 40 29 43 27 98 2
Vindesine 3
or
Cisplatin 40
Doxorubicin 40
Cyclophosphamide 400
Southampton17 1983-1986 32 Cisplatin 120 6 0 0 100 0 81 1 50 34 15 91 9
Vinblastine 3 15
NRH19 1983-1987 87 Cisplatin 70 4 0 2 0 48 60 40 40 39 21 89 11
Etoposide 100
20
Ancona 1 1985-1988 128 Cisplatin 80 P/T 0 0 41 59 88 12 NK NK NK 95 5
Cyclophosphamide 500
Epirubicin 50
alternating with
Methotrexate 30
Etoposide 200
Lomustine 70 (orally)
CEP-8522 1985-1988 49 Cisplatin 120 8 0 2 0 94 67 27 47 33 14 84 16
Vindesine 3 18
UCLA23 1984-1986 63 Cisplatin 120 P/T 0 0 0 100 67 10 40 54 6 63 11
Vinblastine 6
JLCSG25 1990-1995 48 Carboplatin 300 8 0ⴱ 0ⴱ 9ⴱ 91ⴱ NK NK NK NK NK 76 23
Etoposide 120 ⫻ 2
BLT110 1995-2001 477 Cisplatin 80 3 3 20 36 38 82 25 56 24 25 76 31
Vindesine 3⫻2
or
Cisplatin 80 3
Vinorelbine 30 ⫻ 2
or
Cisplatin 50 3
Mitomycin C 6
Vinblastine 6
Using other platinum
regimen
AOI-Udine21 1984-1986 102 Cisplatin 75 6 0 0 0 100 49 51 48 35 17 98 2
Mitomycin C 10
Cyclophophamide 400
MIC224 1988-1996 359 Cisplatin 50 4 NK NK NK NK 64 28 56 27 19 84 19
Mitomycin C 6
Ifosfamide 3 (g/m2)
BLT210 1995-2001 248 Cisplatin 50 3 2 30 30 36 71 19 46 22 19 56 33
Mitomycin C 6
Ifosfamide 3 (g/m2)
Using vinca-alkaloid/
etoposide only
Gwent 226 1982-1984 186 Etoposide 600 6 24 12 10 33 87 8 100 0 0 84 13
ELVIS27 1996-1997 161 Vinorelbine 30 ⫻ 2 6 0 1 15 60 63 21 36 29 26 ⬍1 84
Using anti-metabolic
agent only
Manchester 128 1994-1996 300 Gemcitabine 1,000 ⫻ 3 6 NK NK NK 40ⴱ 94 6 NK NK NK 69 31
Using taxane only
Manchester 229 1995-1997 157 Paclitaxel 200 P/T 0 0 45 55 81 19 47 29 20 66 34
Histology data were provided for 15 trials and performance status and Overall Survival
stage were supplied for 13 trials. Based on these available data, patients Survival analyses were based on 2,533 deaths and 2,714 patients
were mostly male, age between 60 and 70 years, with good perfor- from 16 trials. Figure 1 shows a highly statistically significant benefit of
mance status. Performance status was defined as good (WHO/Eastern chemotherapy on survival (HR, 0.77; 95% CI, 0.71 to 0.83; P ⬍ .0001)
Cooperative Oncology Group 0 or 1, Karnofsky 100 to 70) or poor translating to an absolute improvement of 9% at 12 months increasing
(WHO/Eastern Cooperative Oncology Group 2⫹, Karnofsky 60 survival from 20% to 29% or an absolute increase in median survival
or lower). of 1.5 months (from 4.5 months to 6 months). There was some
Of the stage data we received, 90% of patients had tumors that evidence of heterogeneity between the trials (P ⫽ .02; I2 ⫽ 47%).
were advanced (predominantly stage IIIb and IV). However, there was However, repeating the sensitivity analysis carried out in 1995,6 which
a small proportion (3%) of patients had stage I and II disease. This excluded the extreme results of CEP-8522 (49 patients) resulted in
appears to be because some trials10,26 did not restrict entry to advanced considerably lower heterogeneity (P ⫽ .275; I2 ⫽ 16%) with a similar
patients and these individuals were (presumably) randomly assigned effect of chemotherapy on overall survival (HR, 0.78; 95% CI, 0.72 to
because their condition precluded or the patient declined surgery or 0.85). Also, based on all trials, results using the random effects model
radical radiation therapy. were similar (HR, 0.75; 95% CI, 0.67 to 0.84; P ⬍ .0001). The survival
Most patients had squamous cell tumors (43%) or adenocarci- curve is shown in Figure 2.
nomas (23%), the proportion of these tumor types has not substan- There was no clear evidence of a difference in the effect of chem-
tially changed between 1995 and the current analysis. otherapy between chemotherapy types (Fig 1; P ⫽ .63) or between
The median follow-up for all surviving patients was 1 year 4 trials that used combination chemotherapy and those that used single
months (range, ⬍ 1 month to 9.5 years). agent chemotherapy (Table 3; P ⫽ .40).
SC + CT SC alone
Trial Id. (no. events/no. entered) Hazard Ratio (fixed) 95% CI P value
Fig 1. Hazard ratio plot of effect of chemotherapy on survival. Each trial is represented by a blue square, the center of which denotes the hazard ratio for that trial
with the horizontal lines showing the 99% and 95% CIs. The size of the square is directly proportional to amount of information in the trial. The red diamond gives the
overall hazard ratio for combined results of all trials; the center denotes the hazard ratio and the extremities the 95% CI. The yellow diamonds represent hazard ratios
for the trial groups; the center denotes the hazard ratio and the extremities the 95% CI. SC, supportive care; CT, chemotherapy; NCIC CTG, National Cancer Institute
of Canada Clinical Trials Group; NRH, Norwegian Radium Hospital; CEP-85, Cerce d’etudes pneumologiques; UCLA, University of California—Los Angeles; JLCSG, Joint
Lung Cancer Study Group; BLT, Big Lung Trial; AOI, Associazione Oncologia Italiana; MIC2, mitomycin, ifosfamide, and cisplatin; ELVIS, Elderly Lung Cancer Vinorebine
Italian Study Group.
Of the three trials that we could not include in these analyses, we noma or squamous cell histology. Despite these differences, there
could estimate a HR for survival30 for one trial16 of 207 patients. This was no evidence of a difference in effect of chemotherapy between
trial used single-agent docetaxel, had a reported P ⫽ .03 and gave a trials in the 1995 analysis and the recent trials (interaction P ⫽ .77)
very similar result (HR, 0.70; 95% CI, 0.51 to 0.95) to the included or between previous platinum-based trials and recent platinum-
trial29 that used a single-agent taxane (HR, 0.69; 95% CI, 0.49 to 0.97). based trials (interaction P ⫽ .64) (Table 3).
Since the 1995 meta-analysis, the patient demographic may
have changed, and so the effect of chemotherapy may also have Patient Subgroups
changed. Certainly, there was a higher proportion of patients older There was no clear evidence of a difference or trend in the relative
than 70 in the more recent trials, probably due to aging populations effect of chemotherapy in patient subgroups defined by age (P ⫽ .64),
and widening eligibility criteria, although all trials included pa- sex (P ⫽ .77), stage (P ⫽ .35), histology (P ⫽ .75), or performance
tients older than 70 years (Table 1). The median age of patients in status (P ⫽ .54; Fig 3). Furthermore, despite the difference in under-
the trials included since the 1995 analysis was higher (66 years old, lying survival by performance status, the absolute effect at 12 months
previously 61 years old), there were more women (28%, previously was fairly similar; 8% for performance status 0/Karnofsky 100 to 90
19%), and far more stage IIIa patients (16%, previously 3%). There (from 26% to 34%), 8% for 1/Karnofsky 80 to 70 (from 18% to 26%),
was no real change in the proportion of patients with adenocarci- and 6% for 2⫹/Karnofsky 60 or lower (from 8% to 14%).
SC+CT SC alone
(no. events/no. entered) Hazard Ratio (fixed)
Age
< 60 429/456 368/382
60-64 275/299 267/289
Trend P = .64
65-69 243/268 253/267
≥ 70 336/365 344/367
Sex
Stage
IV 477/508 451/466
Fig 3. Effect of chemotherapy on sur-
Histology
vival by age, sex, stage, histology, and
performance status.
Adenocarcinoma 302/320 295/300
Performance Status
Exploratory Analysis
Performance Status (WHO or Karnofsky)
2+ or ≤ 60 298/310 279/284
The current meta-analysis suggests that we do not need another quality of life. Therefore, all patients who are fit enough and wish to
trial of supportive care alone versus supportive care and chemother- receive chemotherapy, should do so.
apy. What we do need are more trials comparing third generation
chemotherapy combinations, doses, and duration, and also further
AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS
research on toxicity and adverse effects. Recent trials have shown the OF INTEREST
efficacy of epidermal growth factor receptor tyrosine kinase inhibitors
in patients with advanced cancer35 who have relapsed after first-line Although all authors completed the disclosure declaration, the following
chemotherapy and so future trials may also exploit these encourag- author(s) indicated a financial or other interest that is relevant to the subject
ing developments. matter under consideration in this article. Certain relationships marked
This meta-analysis demonstrates conclusively that chemothera- with a “U” are those for which no compensation was received; those
relationships marked with a “C” were compensated. For a detailed
py increases overall survival for all types of patients with advanced
description of the disclosure categories, or for more information about
NSCLC and that there should be no change in the treatment para- ASCO’s conflict of interest policy, please refer to the Author Disclosure
digm. Evidence from trials that collected quality of life data also sug- Declaration and the Disclosures of Potential Conflicts of Interest section in
gests that this approach is unlikely to be detrimental to the patients’ Information for Contributors.
Employment or Leadership Position: Thierry Le Chevalier, Collection and assembly of data: S. Burdett, L. Stewart
GlaxoSmithKline (U) Consultant or Advisory Role: None Stock Data analysis and interpretation: R. Arriagada, A. Auperin, D. Bell,
Ownership: None Honoraria: Richard Stephens, Pierre Fabre Oncology S. Burdett, G. Cartei, Y. Cormier, M. Cullen, P. Ganz, C. Gridelli,
Research Funding: Max Parmar, UK Medical Research Council; J. Higgins, D. Johnson, S. Kaasa, T. Le Chevalier, C. Le Pechoux,
Jean-Pierre Pignon, Sanofi-aventis; Richard Stephens, UK Medical M. Parmar, J.P. Pignon, E. Quoix, E. Rapp, L. Seymour, R. Souhami,
Research Council Expert Testimony: None Other Remuneration: None S. Spiro, R. Stephens, L. Stewart, N. Thatcher, J. Tierney, D.
Tummarello, J. van Meerbeeck, C. Williams, I. Williamson
AUTHOR CONTRIBUTIONS Manuscript writing: R. Arriagada, A. Auperin, S. Burdett, J. Higgins,
D. Johnson, T. Le Chevalier, C. Le Pechoux, M. Parmar, J.P. Pignon,
Conception and design: R. Arriagada, A. Auperin, S. Burdett, J. Higgins, R. Souhami, R. Stephens, L. Stewart, J. Tierney, J. van Meerbeeck
D. Johnson, T. Le Chevalier, C. Le Pechoux, M. Parmar, J.P. Pignon, Final approval of manuscript: R. Arriagada, A. Auperin, D. Bell,
R. Souhami, R. Stephens, L. Stewart, J. Tierney, J.P. van Meerbeeck S. Burdett, G. Cartei, Y. Cormier, M. Cullen, P. Ganz, C. Gridelli,
Administrative support: S. Burdett J. Higgins, D. Johnson, S. Kaasa, T. Le Chevalier, C. Le Pechoux,
Provision of study materials or patients: D. Bell, G. Cartei, Y. Cormier, M. Parmar, J.P. Pignon, E. Quoix, E. Rapp, L. Seymour, R. Souhami,
M. Cullen, P. Ganz, C. Gridelli, S. Kaasa, E. Quoix, E. Rapp, L. Seymour, S. Spiro, S. Spiro, R. Stephens, L. Stewart, N. Thatcher, J. Tierney, D.
R. Stephens, N. Thatcher, D. Tummarello, C. Williams, I. Williamson Tummarello, J. van Meerbeeck, C. Williams, I. Williamson
cancer: Best supportive care (BSC) versus BSC plus 25. Helsing M, Bergman B, Thaning L, et al:
REFERENCES chemotherapy. Lung Cancer 24:17-24, 1999 Quality of life and survival in patients with advanced
15. Gasparini S, Farabollini B, Isidori P, et al: non-small cell lung cancer receiving supportive care
1. Parkin DM, Bray F, Ferlay J, et al: Global Lonidamine, chemotherapy (cisplatin and etoposide) plus chemotherapy with carboplatin and etoposide
cancer statistics, 2002. CA Cancer J Clin 55:74-108, and chemotherapy plus lonidamine vs the best or supportive care only. A multicentre randomised
2005 supportive care in the treatment of advanced non- phase III trial. Eur J Cancer 34:1036-1044, 1998
2. Visbal AL, Leighl NB, Feld R, et al: Adjuvant small cell lung cancer (NSCLC): Preliminary results 26. Anderson G, Payne H: Response rate and
chemotherapy for early stage non-small cell lung of a randomised study. Lung Cancer 7:A383, 1991 toxicity of etoposide (VP-16) in squamous carcinoma
cancer. Chest 128:2933-2943, 2005 16. Roszkowski K, Pluzanska A, Krzakowski M, et of the lung: Report from the Lung Cancer Treatment
3. World Health Organization: Global cancer rates al: A multicenter, randomized, phase III study of Study Group. Semin Oncol 12:21-22, 1985 (suppl 2)
could increase by 50% to 15 million by 2020. http:// docetaxel plus best supportive care versus best 27. Elderly Lung Cancer Vinorelbine Italian Study
www . who . int / mediacentre / news / releases/2003/ supportive care in chemotherapy-naı̈ve patients with Group: Effect of vinorelbine on quality of life and
pr27/en/ metastatic or non-resectable localized non-small cell survival of elderly patients with advanced non-small
4. Greenlee RT, Murray T, Bolden S, et al: Can- lung cancer. Lung Cancer 27:145-157, 2000 cell lung cancer. J Natl Cancer Inst 91:66-72, 1999
cer Statistics, 2000. CA Cancer J Clin 50:7-33, 2000 17. Woods RL, Williams CJ, Levi J, et al: A 28. Anderson H, Hopwood P, Stephens RJ, et al:
5. Rudd R: Chemotherapy in the treatment of randomised trial of cisplatin and vindesine versus Gemcitabine plus best supportive care (BSC) vs BSC
non-small cell lung cancer. Respiratory Disease in supportive care only in advanced non-small cell lung in inoperable non-small cell lung cancer: A random-
Practice 7:12-14, 1991 cancer. Br J Cancer 61:608-611, 1990 ized trial with quality of life as the primary outcome.
6. Non-Small Cell Lung Cancer Collaborative 18. Rapp E, Pater JL, Willan A, et al: Chemother- Br J Cancer 83: 447-453, 2000
Group: Chemotherapy in non-small cell lung cancer: apy can prolong survival in patients with advanced 29. Ranson M, Davidson N, Nicolson M, et al:
A meta-analysis using updated data on individual non-small cell lung cancer: Report of a Canadian Randomized trial of paclitaxel plus supportive care
patients from 52 randomised clinical trials. BMJ multicenter randomized trial. J Clin Oncol 6:633-641, versus supportive care for patients with advanced
311:899-909, 1995 1988 non-small cell lung cancer. J Natl Cancer Inst 92:
7. Burdett S, Pignon JP, Stewart L: Chemother- 19. Kaasa S, Lund E, Thorud E, et al: Symptomatic 1074-1080, 2000
apy in non-small cell lung cancer: An update of an treatment versus combination chemotherapy for 30. Parmar MKB, Torri V, Stewart L: Extracting
IPD meta-analysis. J Thorac Cardiovasc Surg 129: patients with extensive non-small cell lung cancer. summary statistics to perform meta-analyses of the
1205, 2005 Cancer 67:2443-2447, 1991 published literature for survival endpoints. Stat Med
8. Dickersin K, Scherer R, Lefebvre C: Identify- 20. Cellerino R, Tummarello D, Guidi F, et al: A 17:2815-2834, 1998
ing relevant studies for systematic reviews. BMJ randomized trial of alternating chemotherapy versus 31. Lilenbaum RC, Langenberg P, Dickersin K:
309:1286-1291, 1994 best supportive care in advanced non-small cell lung Single agent versus combination chemotherapy in
9. Yusuf S, Peto R, Lewis J, et al: Beta blockade cancer. J Clin Oncol 9:1453-1461, 1991 patients with advanced non-small cell lung carci-
during and after myocardial infarction: An overview 21. Cartei G, Cartei F, Cantone A, et al: Cisplatin- noma. Cancer 82:116-126, 1998
of the randomized trials. Prog Cardiovasc Dis 27: cyclophosphamide-mitomycin combination chemo- 32. Hotta K, Matsuo K, Ueoka H, et al: Addition of
335-371, 1985 therapy with supportive care versus supportive care platinum compounds to a new agent in patients with
10. Spiro SG, Rudd RM, Souhami RL, et al: Chem- alone for treatment of metastatic non-small cell lung advanced non-small cell lung cancer: A literature
otherapy versus supportive care in advanced non- cancer. J Natl Cancer Inst 85:794-800, 1993 based meta-analysis of randomised trials. Ann Oncol
small cell lung cancer: Improved survival without 22. Quoix É, Dietemann A, Charbonneau J, et al: 15:1782-1789, 2004
detriment to quality of life. Thorax 59:828-836, 2004 Is cisplatinum-based chemotherapy useful in dis- 33. Delbaldo C, Michiels S, Syz N, et al: Benefits
11. Parmar MKB, Machin D: Survival Analysis: A seminated non–small-cell lung cancer? Report of a of adding a drug to a single-agent or a 2-agent
Practical Approach. Chichester, John Wiley and French multicenter randomized trial [in French]. Bull chemotherapy regimen in advanced non-small cell
Sons Ltd, 1995 Cancer 78:341-346, 1991 lung cancer. JAMA 292:470-484, 2004
12. Higgins JPT, Thompson SG, Deeks JJ, et al: 23. Ganz PA, Figlin RA, Haskell CM, et al: Sup- 34. Lilenbaum RC, Herndon JE II, List MA, et al:
Measuring inconsistency in meta-analyses. BMJ portive care versus supportive care and combination Single agent versus combination chemotherapy in
327:557-560, 2003 chemotherapy in metastatic non-small cell lung can- advanced non-small cell lung cancer: The Cancer
13. Kaplan EL, Meier P: Nonparametric estima- cer. Does chemotherapy make a difference? Cancer and Leukemia Group B (study 9730). J Clin Oncol
tion from incomplete observation. Journal of the 63:1271-1278, 1989 23:190 196, 2005
American Statistical Association 53:457-481, 1958 24. Cullen MH, Billingham LJ, Woodroffe CM, et 35. Gridelli C, Bareschino MA, Schettino C, et al:
14. Thongprasert S, Sanguanmitra P, Juthapan al: Mitomycin, ifosfamide and cisplatin in resectable Erlotinib in non-small cell lung cancer treatment:
W, et al: Relationship between quality of life and non-small cell lung: Effects on survival and quality of Current status and future development. The Oncol-
clinical outcomes in advanced non-small cell lung life. J Clin Oncol 17:3188-3194, 1999 ogist 12:840-849, 2007
■ ■ ■
Acknowledgment
The NSCLC Meta-Analyses Collaborative Group (composition below) thanks all patients who took part in the trials and contributed to this
research. The meta-analysis would not have been possible without their participation or without the collaborating institutions that provided their
trial data.
The meta-analysis was funded by the United Kingdom Medical Research Council.
NSCLC Meta-Analyses Collaborative Group
Project Management Group
Medical Research Council (MRC) Clinical Trials Unit, London, United Kingdom
S. Burdett, R. Stephens, L. Stewart, J. Tierney
Institut Gustave-Roussy, Villejuif, France
A. Auperin, T. Le Chevalier, C. Le Pechoux, J.P. Pignon
International Advisory Group
R. Arriagada, Karolinska Institutet, Stockholm, Sweden and Institut Gustave-Roussy, Villejuif, France
J. Higgins, MRC Biostatistics Unit, Cambridge, United Kingdom
D. Johnson, Vanderbilt-Ingram Cancer Center, Nashville, TN
J. van Meerbeeck, University Hospital, Ghent, Belgium
M. Parmar, MRC Clinical Trials Unit, London, United Kingdom
R. Souhami, Cancer Research United Kingdom, London, United Kingdom
Collaborators who supplied individual patient data
D. Bell, Royal North Shore Hospital, St Leonards, New South Wales, Australia
G. Cartei, International Academy of Environmental Sciences, Venice, Italy
Y. Cormier, Hospital Laval, Sainte-Foy, Quebec, Canada
M. Cullen, Queen Elizabeth Hospital, Birmingham, United Kingdom
P. Ganz, Schools of Medicine and Public Health, University of California, Los Angeles, Los Angeles, CA
C. Gridelli, SG Moscati Hospital, Avellino, Italy
S. Kaasa, University Hospital, Trondheim, Norway
E. Quoix, Service de Pneumologie, Strasbourg, France
E. Rapp (Retired), University of California-Los Angeles, Tom Baker Cancer Centre, Calgary, Alberta, Canada
L. Seymour, National Cancer Institute of Canada Clinical Trials Group, Queens University, Kingston, Ontario, Canada
S. Spiro, University College Hospital, London, United Kingdom
N. Thatcher, Christie Hospital, Manchester, United Kingdom
D. Tummarello, Universita Delgi Studi de Ancona, Ancona, Italy
C. Williams, Bristol Hematology and Oncology Centre, Bristol, United Kingdom
I. Williamson, Newport Chest Clinic, Newport, United Kingdom